Ontology based web Crawling - A novel approach

Abstract

The requirement of a web Crawler that downloads most relevant pages is still a major challenge in the field of Information Retrieval Systems. The use of link analysis algorithms as page rank and other Importance-metrics like back link count have shed a new approach in prioritizing the URL queue for downloading higher relevant pages. In this paper, the combination of these metrics along with a new metric called association-metric, which brings the use of Ontology for Crawling has been proposed and implemented. The use of domain dependent Ontology brings into effect the both semantic and link nature of the URL and its page.The association-metric estimates the semantic content of the URL based on the domain dependent ontology, which in turn strengthens the metric that is used for prioritizing the URL queue. In addition, after downloading the page, the association metric plays important role in estimating the relevancy of the links in that page. This new metric solves the major problem of finding the relevancy of the pages before the process of crawling, to an optimal level. The crawler developed based on the Association metric has shown encouraging results. © Springer-Verlag Berlin Heidelberg 2005.